Comparison of Deep Learning Approaches for Multi-Label Chest X-Ray Classification
Comparison of Deep Learning Approaches for Multi-Label Chest X-Ray Classification
复制标题
DOI：
10.1038/s41598-019-42294-8
复制
发表时间：
2019-04-23
影响因子：
4.6
通讯作者：
Saalbach, Axel
中科院分区：
文献类型：
--
作者：
Baltruschat, Ivo M.;Nickisch, Hannes;Saalbach, Axel
随着标记的X射线图像档案(例如胸片X-射线数据集)可用性的增加，人们对深度学习技术的兴趣与日俱增。为了更好地了解不同的方法及其在胸部X光分类中的应用，我们详细研究了一个强大的网络架构：ResNet-50。在这一领域以往工作的基础上，我们考虑在微调和不微调的情况下进行转移学习，以及从零开始培训专用X射线网络。为了利用X射线数据的高空间分辨率，我们还包括扩展的ResNet-50架构，以及在分类过程中整合非图像数据(患者年龄、性别和采集类型)的网络。在总结实验中，我们还调查了多个ResNet深度(即ResNet-38和ResNet-101)。在系统评估中，我们使用5倍重采样和多标签损失函数，通过ROC统计量比较不同的病理分类方法的性能，并使用等级相关分析分类器之间的差异。总体而言，我们观察到在所实现的性能中的相当大的分布，并得出结论，X光特定的ResNet-38，整合非图像数据产生了最好的整体结果。此外，使用类激活图来理解分类过程，并详细分析了非图像特征的影响。
The increased availability of labeled X-ray image archives (e.g. ChestX-ray14 dataset) has triggered a growing interest in deep learning techniques. To provide better insight into the different approaches, and their applications to chest X-ray classification, we investigate a powerful network architecture in detail: the ResNet-50. Building on prior work in this domain, we consider transfer learning with and without fine-tuning as well as the training of a dedicated X-ray network from scratch. To leverage the high spatial resolution of X-ray data, we also include an extended ResNet-50 architecture, and a network integrating non-image data (patient age, gender and acquisition type) in the classification process. In a concluding experiment, we also investigate multiple ResNet depths (i.e. ResNet-38 and ResNet-101). In a systematic evaluation, using 5-fold re-sampling and a multi-label loss function, we compare the performance of the different approaches for pathology classification by ROC statistics and analyze differences between the classifiers using rank correlation. Overall, we observe a considerable spread in the achieved performance and conclude that the X-ray-specific ResNet-38, integrating non-image data yields the best overall results. Furthermore, class activation maps are used to understand the classification process, and a detailed analysis of the impact of non-image features is provided.